06-16-2013 03:52 AM - edited 03-04-2019 08:13 PM
Dears,
I have two N7ks 6.1(2) on my network and 10 access switches are connected to this N7Ks via VPC(10G on each link).
We have faced a problem. By mistakenly somebody wrongly connected a cable from access switch1 to access switch2.
Immediately the entire network went down and some access switch portchannel(VPC) also went down
Real Time Log is attached.I doubt something wrongly went on Nexus
It would be really great if someone can help me on this
Regards
Mohd
Solved! Go to Solution.
06-16-2013 10:49 PM
Duplicate posts.
06-16-2013 08:43 PM
This should probably be in the datacenter community. Can you post the interface configs?
Sent from Cisco Technical Support Android App
06-16-2013 10:38 PM
Hi ,
I attached all the logs required ( Uplinks configurations, logs during the problem ) .
Please let me again discuss the issue . We have two nexus working as core switches perfectly and we have alot of access switches connect to them , by mistake two cables where connected from access switch 1 to access switch 2 and native vlan mismatch happened suddenly and bring all other access switches down . I attached all the uplinks for nexus switches , switch 1 , switch 2 , & switch 3 (one of the access switches that was affected by the loop created on switch 1 & 2)
06-16-2013 10:49 PM
Duplicate posts.
06-16-2013 11:34 PM
Ignore this discussion and please help the same on Data center discussion .
Thanks in advance
07-02-2013 12:37 AM
This question has been answered on it in the data center discussion after mistakenly duplicate post and the correct answer was :
The problem you saw is due to the spanning-tree bpdufilter enable command you have enabled on these interfaces.
This command means the switch will filter any BPDU it receives on the interface, and so prevent the Catalyst 6500 detecting that the port is connected to another switch. As soon as the ports went into a forwarding state you had a loop through which broadcasts etc., would have circulated continuously. Any broadcasts received on these ports would have also been forwarded on the port-channel to the Nexus 7000.
If these are edge ports, as would seem to be the case as you have spanning-tree portfast edge configured on two of them, then you should not enable spanning tree BPDU filtering. The consequences are exactly as you have seen when a port is patched incorrectly.
Regards
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide